Abstract
Scholars and community members have long utilized participatory action research (PAR) to strategize local efforts to address real-world problems across various disciplines, including education. Applied linguistics and language education have also embraced and echoed the PAR tenets that situate the locus of knowledge production within local communities, aiming to co-construct ethical praxis for socially just change. However, there are few methodological syntheses of PAR studies in additional language education, despite ongoing calls to promote such a democratic research methodology to effect societal change. Aiming to bring greater methodological clarity regarding PAR in language education research, this article reviews methodological issues, synthesizing PAR in theory and PAR in practice. Drawing on a methodological review approach, it analyzes 23 PAR studies published between 2014 and 2024 that involved teacher participants in additional language education. The review focuses on how the three components of ‘participation,’ ‘action,’ and ‘research’ are framed and represented in these studies. Findings highlight various methodological discussion points in PAR in practice, such as the orientation of action, group formation, naming practices and authorship, positionality statements, participant engagement in data analysis, and methodological frameworks. Based on the findings, the review proposes a set of reflective questions as starting points for dialogue among PAR practitioners. These questions will help explore ways to make PAR more participatory, and in turn, co-construct ethical and impactful PAR for social change.
Keywords
1. Introduction
Participatory action research (PAR) is a methodology that combines research with community practice to transform the status quo toward a more just world, centering participants’ roles in the knowledge creation process (e.g. Fine & Torre, 2021; Kemmis et al., 2014; Lenette, 2022; McIntyre, 2008). As the name suggests, it is the three constructs – participation, action, and research – that serve as the methodological pillars and make PAR unique from other forms of social science research. PAR locates its philosophical roots in anticolonial political movements in the Global South during the 1960s and 1970s (Kapoor & Jordan, 2009; Lenette, 2022; Zavala, 2013); for example, it is associated with the thinking of Brazilian educational philosopher Paulo Freire, a line of thought developed as critical pedagogy. Critical pedagogy aims to emancipate the marginalized by raising their critical awareness of oppression and by transforming power structures that perpetuate social inequalities. For Freire, pedagogy is ‘forged with, not for, the oppressed (whether individuals or peoples) in the incessant struggle to regain their humanity’ (Freire, 1970/2017, p. 22). Aligned with Freire’s vision of participation and action, Colombian sociologist Orlando Fals-Borda also emphasized the need to decenter researchers’ roles in research, arguing, ‘Do not monopolise your knowledge nor impose arrogantly your techniques, but respect and combine your skills with the knowledge of the researched or grassroots communities, taking them as full partners and co-researchers’ (Fals-Borda, 1995, as cited in Chevalier & Buckles, 2019, p. 48). Focusing on doing research with rather than on or for participants, PAR reshapes the way we engage in research, with the primary goal of improving the world by changing it (McIntyre, 2008). This is why PAR is often conceptualized not merely as a research methodology but rather as an epistemology (Kapoor & Jordan, 2009; Lenette, 2022). PAR, as an epistemology, ‘signals a distinctive way of thinking about who has knowledge, who holds expertise, and how new knowledge can be produced, across differences, when the perspective of those most impacted by injustice [is] privileged and fueling movements for change are prioritized’ (Fine & Torre, 2021, p. 10).
In the field of language education research, there is a growing call to adopt this emancipatory and transformative research methodology, as it addresses real-world social issues that language learners need to grapple with, such as racism, linguistic discrimination, ethnic stereotyping, homophobia, sexism, and ableism. The focus on praxis and activism, rather than theoretical discussions in the academic bubble, has thus been emphasized in the field (Avineri & Martinez, 2021; Kubota, 2023; Linville & Whiting, 2020). Indeed, it has been over three decades since TESOL Quarterly published an article on PAR methodology, highlighting its philosophical tenets and the importance of breaking the dichotomy between the researcher and the researched in achieving social justice (Auerbach, 1994). Since then, the value of PAR in theory has been recognized and has occasionally come into the spotlight in language education (e.g. Burns, 2013, 2019; Filipović, 2019; Pitura, 2025). However, the use of PAR in this field remains limited despite the transformative potential of the methodology. PAR in theory has rarely been examined and discussed in tandem with PAR in practice; in particular, there has been a lack of methodological synthesis that analyzes how scholars and participants in language education have interpreted, practiced, and documented PAR. This has prevented scholars and community members from having conceptual and methodological clarity and, in turn, from realizing its potential. For some, PAR might be simply understood as a ‘participatory’ version of action research, but this understanding is not conceptually useful since any action research in language education inherently entails some form of participation by research participants (e.g. teachers, students). Realizing its methodological potential requires deep engagement with the semantic ambiguity and implications underlying each letter in the acronym PAR; we must have a clearer understanding of what we mean by participation, action, and research.
In response to the need outlined above, this review aims to bring greater methodological clarity regarding PAR in language education research, highlighting key considerations for enacting its methodological tenets. Drawing on a methodological review approach (Chong & Plonsky, 2024), it analyzes 23 PAR studies conducted by researchers and teacher participants in the context of additional language education, with focused attention on how the three components of ‘participation,’ ‘action,’ and ‘research’ are framed and represented in their PAR. Based on the findings, the review proposes a set of reflective questions as starting points for dialogue among PAR practitioners, which will help explore ways to make PAR more participatory and, in turn, to co-construct ethical and impactful PAR for social change.
2. Methodology
Methodological reviews, a form of secondary research synthesis, assess and summarize methodological issues and practices in research within a specific domain (Chong & Reinders, 2021). They are conducted systematically and documented according to the structure of typical empirical studies in the social sciences (Chong & Plonsky, 2024), although a non-systematic, narrative approach is also possible (e.g. Lin, 2014). Following Chong and Plonsky (2024), this methodological review adopts a systematic approach, using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram (Page et al., 2021) to document the literature search process and to enhance methodological transparency. This review examines how PAR with teachers is practiced and documented in additional language education contexts, focusing on the tenets of PAR encapsulated in its acronym. The research question is:
How are the three components of participation, action, and research framed and represented in PAR studies with teachers in additional language education?
2.1. Literature search
Based on the research question, the inclusion criteria for this methodological review were set as follows: (1) PAR with teachers, including pre-service teachers, (2) PAR in additional language education contexts (e.g. second/foreign language education and bilingual education), (3) PAR published between 2014 and 2024, and (4) peer-reviewed journal articles. The literature search was conducted using three major databases to cover a wide range of literature in language education: Educational Resources Information Center (ERIC), JSTOR, and Linguistics and Language Behavior Abstracts (LLBA). The search string used for the identification stage was ‘participatory action research’ AND (‘foreign language’ OR ‘second language’ OR L2 OR bilingual OR ‘language learning’ OR ‘language teaching’) and was applied to all fields, including titles, abstracts, main text, and references. The search filters on the databases were used to identify peer-reviewed articles published between January 1, 2014 and December 31, 2024. 1 The identified articles were imported into a systematic review tool, Covidence, to facilitate the review process.
Figure 1 portrays the study selection process and reasons for exclusion in the PRISMA 2020 flow diagram (Page et al., 2021). The total number of articles yielded for the identification stage was 376. Duplicate entries were removed, after which 346 articles remained. The first screening stage focused on scanning titles and abstract content, and excluded 259 articles that did not meet the inclusion criteria. The remaining 87 articles underwent full-text screening. I excluded 64 articles through this screening, leaving 23 studies eligible for the methodological review (for a list of included articles, see Table 1).
PRISMA flow diagram of study selection process.
List of studies included in this review (n = 23).
2.2. Data extraction and coding procedures
Data were extracted based on a coding scheme (Table 2). The first two coding categories pertain to bibliographic information and contextual characteristics of each article. The remaining coding categories and items were designed to extract key findings addressing the research question. Coding items were deductively and inductively created. Coding was conducted using MAXQDA, a qualitative data analysis software, and the coding results were exported to and organized in Microsoft Excel. In what follows, I detail the coding procedures used to extract data for answering the research question.
Coding scheme: Categories and items.
2.2.1. Coding for the ‘action’ component
To extract data to examine the orientation and characteristics of action in PAR, these items were coded: types of action, areas of inquiry, and criticality. Types of action in PAR refer to the kinds of activities or changes the project reportedly aimed to achieve, and the study’s objectives or goals were coded accordingly. As McIntyre (2008) notes, actions are ultimately ‘the result of the questions [participants] pose, examine, and address’ within the PAR process. Thus, areas of inquiry (e.g. research questions) were also coded to better understand the study’s action component. Lastly, the criticality of the action taken in the project was examined. This decision was motivated by the consensus among PAR theorists and practitioners that PAR is a methodological framework used to transform the status quo toward a more just world, principles of which are rooted in multiple ‘critical’ traditions and paradigms such as critical theory, critical pedagogy, and decolonial thought (Fine & Torre, 2021; Jacobs, 2018; Kapoor & Jordan, 2009; Kemmis & McTaggart, 2007; Kincheloe, 2009; Lenette, 2022). This item was first open-coded and categorized according to the type of action, and then coded with three options: critical, non-critical, or potentially critical. This coding was based on my interpretative reading of the entire content of each article, particularly focusing on PAR objectives, problem statements, research questions, research designs, literature reviews, and discussion sections. I recognize that this reading was indeed a challenging task, and it was even presumptuous to assume that I could judge this quality, given the fact that the construct of criticality itself is under scrutiny (Kubota & Miller, 2017; Pennycook, 2021) and that its definition is inherently diverse and context-dependent (Crookes, 2021). Kubota and Miller (2017) caution that research topics such as identity, race, gender, and class, per se, do not lend themselves to criticality. An in-depth analysis of ‘power, inequalities, domination, and resistance,’ combined with ‘praxis with hyper self-reflexivity and result-oriented action’ (p. 148), is central to the critical. Informed by this, the guiding criteria for this coding were whether the projects explicitly or implicitly drew on constructs such as power, domination, oppression, inequality, equity, inclusion, critical pedagogy, emancipation, decolonization, and so forth. I did not examine whether these constructs were simply mentioned in the articles; rather, my focus was on whether the ideas were there and informed the projects.
2.2.2. Coding for the ‘participatory’ component
Issues around the construct of participation are central to methodological discussions on PAR (Chevalier & Buckles, 2019; Kindon et al., 2007). In PAR, participants are not research subjects to be studied, but research agents who actively engage in the fluid, spiral, and often overlapping process of planning, acting, observing, reflecting, and then replanning for further actions (Kemmis et al., 2014; McIntyre, 2008; McTaggart, 1997). This is not just a research method but speaks to its central ideological commitment to ‘disrupt conventional hierarchies of knowledge production: who decides on the questions to ask, how to ask them and how to theorise the world’ (Walker & Boni, 2020, p. 15). The coding items of the ‘participatory’ component were thus created based on the view that participants’ ownership of research and practice, which frames their roles as central to scientific inquiry, is key to conceptualizing participation in PAR (Auerbach, 1994; Fine & Torre, 2021; McTaggart, 1997). Accordingly, the coding of the following items aimed to examine how teacher participants were positioned in PAR: group formation, naming practices, data analysis involvement, positionality statements, and participant authorship. Group formation refers to how PAR groups were formed and how participants started or joined the projects. Naming practices concern how teacher participants are identified and represented in the reporting of PAR. The coding of data analysis involvement aimed to identify if and how participants engaged in the analysis and interpretation of data that form the basis of the findings presented in the PAR studies. Positionality statements were coded to examine whether the studies reported teacher participants’ social positioning, identity, and background. Finally, participant authorship was examined and coded based on the assumption that the inclusion of participants as co-authors is a strong indicator of research ownership and participant contribution.
2.2.3. Coding for the ‘research’ component
The action and participatory components are the driving forces of PAR and receive much attention in its methodological discussions; however, PAR also entails research. To extract data on how research is conceptualized in PAR, I decided to code the PAR theories that inform the studies, the data used in the studies, and the data analysis methods. Thus, the coding items were methodological framing, types of data, and data analysis frameworks/methods, which were primarily identified in the methodology sections of the studies. The construction of these coding items was motivated by an exploration of alternative conceptualizations of research, which aim to disrupt the research-action dualism (Kemmis & McTaggart, 2007). By coding these items, the analysis sought to identify how PAR epistemology, which asks ‘who has knowledge, who holds expertise, and how new knowledge can be produced’ (Fine & Torre, 2021, p. 10), is enacted methodologically. Coding methodological framing focused on in-text citations of PAR theories, aiming to understand how the PAR studies were theoretically framed and what methodological premises and tenets appeared to underlie their research. Coding for types of data identified the data sources used in the studies, examining the types of data commonly used in PAR. Finally, the coding of data analysis frameworks/methods examined how the data were analyzed, coding the analytical framework and approach that informed the analysis. Thus, in-text citations of data analysis literature were a key focus in coding.
2.3. Trustworthiness and limitations
This methodological review incorporated a systematic literature synthesis process drawing on the PRISMA guidelines (Page et al., 2021), aiming to improve methodological clarity and transparency by detailing the study selection and data extraction processes. Along with this procedural strategy, I recognize that trustworthiness can also be enhanced when reviewers exercise reflexivity about how their positioning and paradigmatic orientations influence their research synthesis (Hampson & McKinley, 2024; Maeda et al., 2022). It is thus important to explain my researcher positionality. I am a PhD candidate doing a PAR dissertation with high school English teachers in Japan, and my motivation for conducting this review was tied to my desire to learn both PAR in theory and PAR in practice for my dissertation project. Like the teachers I work with, I was also a high school teacher in Japan, where teachers have often been treated as the ‘researched’ or knowledge receivers rather than researchers and knowledge producers. Believing that teacherly practice is conducive to locally situated knowledge production, I participate in the academic community with a strong sense of practitioner identity. I consider myself a teacher-researcher or pracademic with a praxis-focused approach to educational research that resists positivism and scientism, which feign neutrality and compartmentalize the roles of theorizers (researchers) and practitioners (teachers). I hold that a depoliticized skill-based language education backed by the logic of neutrality, standardization, and accountability runs the risk of maintaining and perpetuating normative (language) ideologies (e.g. native-speakerism, heteronormativity, ethnic nationalism, ableism), keeping structural injustice in society intact. This pedagogical concern, along with my identity as a pracademic, forms the basis of my paradigmatic position, and my research has been informed by the critical research tradition which ‘refuses to take knowledge claims at their face value; relentlessly interrogates all bodies of knowledge to unearth the interests shaping them; and is committed to creating knowledge that is explicitly intended for emancipatory purposes’ (Prasad, 2018, p. 159). Like many other PAR theorists and practitioners, I see PAR as epistemologically rooted in this tradition and as providing reflexive spaces to rethink what participation entails and what counts as legitimate scientific knowledge. This methodological positioning influences how I approach this study, as indicated in the description of the coding procedure. Most of the PAR literature cited to provide rationales for creating coding items for this review comes from work situated within the critical research tradition. Consequently, my discussions and appraisals of the results are also drawn from critical perspectives on participatory research.
Before presenting the findings, I should acknowledge that this methodological review is not without limitations. First, this review does not engage with the terminological and methodological variants of PAR. Relying on the search string ‘participatory action research,’ I could not cast a wide net to include studies that might be broadly categorized as PAR but use different terminology, such as community-based participatory research, participatory learning and action, and, more simply, action research. This study reviewed only studies that explicitly use the term PAR to describe and frame their approach, and I note that this represents a methodological limitation of this review, not necessarily my understanding of PAR. I recognize and support an inclusive understanding of PAR that embraces diverse community-driven participatory work (Chevalier & Buckles, 2019; Cornish et al., 2023). Second, this review is limited in that it does not include PAR studies published in languages other than English. This also results from the search string used for the literature search. Additionally, this review only engages with peer-reviewed journal articles, and other forms of knowledge creation and dissemination were not considered for review, although I consider them legitimate outlets of PAR. These limitations reflect pragmatic constraints in conducting a systematic methodological review based on academic databases. I acknowledge that the use of databases and their filters – as reflected in the literature selection process in this review – privileges certain forms of knowledge and marginalizes others, which is the very structure that PAR seeks to disrupt.
Finally, I acknowledge that I was the sole reviewer/coder in this methodological review. The use of multiple researchers is generally considered a recommended practice in research synthesis (Suri, 2013), and I recognize that this review may have taken a different shape if I had been able to work with multiple collaborators from different paradigmatic backgrounds. Yet, I aimed to address this limitation through a reflexive research design and detailed documentation of the review process, as described above.
3. Results
This section first provides the contextual characteristics of the reviewed PAR studies to illustrate overall trends, followed by the main results that answer the research question. Regarding the main results, I begin with the ‘action’ component to first clarify what the reviewed PAR studies were about, and move on to the other components of ‘participation’ and ‘research.’ The findings for each component are presented and discussed according to the coding items, as described in the methodology section.
3.1. PAR settings and trends
Table 3 displays the contextual characteristics of the reviewed PAR studies. Although the United States is the most represented country (n = 9), the studies come from diverse geographical locations, spanning 13 countries (see Figure 2). As shown in Figure 3, the majority of the studies were conducted in K-12 settings (n = 12), followed by post-secondary settings (n = 9) and studies covering multiple levels of education (n = 2). In terms of languages taught (see Figure 4), English as an additional language, such as ESL and EFL, was dominant (n = 14), presumably due to the English search string used for the literature search. Another five studies were framed as bilingual education (e.g. English-French, English-Spanish). Three additional studies addressed Arabic, French, and Spanish language education, respectively, and the last study involved two additional language subjects: ESL and FSL. Teacher group sizes in the reviewed studies (see Figure 5) ranged from one to 205, which can be categorized as small (1–5 teachers, n = 8), medium (8–36 teachers, n = 13), and large (129 and 205 teachers, n = 2) groups.
Contextual characteristics of the reviewed studies.
Notes. a Number indicates graduate student mentors; the number of classroom teachers is unknown. b Based on the initial needs assessment data; the actual participant number is unclear. c Based on the course enrollment; the actual participant number is unclear. d Based on the original dissertation.
Geographical locations of the reviewed studies.
Education levels of the reviewed studies.
Languages taught in the reviewed studies.
Teacher group sizes in the reviewed studies.
3.2. The ‘action’ component
3.2.1. Types of action
Since the reviewed studies are PAR with teachers, it is natural that the actions focused on in their PAR were closely aligned with the participants’ professional responsibility: teaching. Most studies, as presented in Table 4, were categorized as enactment of named pedagogies (n = 15), in which specific pedagogical interventions constituted actions in PAR. The frequently employed pedagogical approaches include photovoice (Call-Cummings et al., 2019; Crawford-Garrett et al., 2020; Villacañas de Castro, 2017), translanguaging (Lau, 2020; Murillo, 2017; Song, 2022), the Freirean approach characterized by the tradition of critical pedagogy (Di Stefano et al., 2021; Ordem, 2023), and critical literacy (Crawford-Garrett et al., 2020). Three studies pertained to the enhancement of assessment practices, such as projects aimed at improving formative assessment (De Neve et al., 2022) and ICT-based assessment (Makruf & Barokah, 2023). One study (López-Gopar et al., 2021) spans both categories, encompassing actions related to pedagogical implementations (e.g. translanguaging) and improving assessment. In two studies, actions were framed broadly as improvement of general teaching practices, such as developing teachers’ confidence in teaching English and their communicative competence (Doungprom et al., 2016) and improving TESL practices by foregrounding teachers’ local knowledge (Modiba & Stewart, 2022). The other two studies were categorized as development of communities and partnerships in teacher education. Coles-Ritchie et al. (2019) aimed to develop a reciprocal campus-community partnership and identify research gaps in campus-community partnerships, while Orsini-Jones’s (2023) focus was on establishing an online ELT community for intercultural teacher education.
Types of action identified in the reviewed studies.
3.2.2. Areas of inquiry
Table 5 shows the areas of inquiry explored in the reviewed studies, categorized based on the coding results of the studies’ research questions or inquiry foci. As the studies often had more than one inquiry focus, the categorization was non-mutually exclusive. Most of the studies were categorized as evaluation of project impact (n = 19), in which the studies presented and discussed their data to evaluate their project outcomes. The studies in documentation of action processes (n = 6) focused on the PAR process, describing in detail how their objective had been achieved. Crawford-Garrett et al. (2020), for example, document how teachers named oppression, deconstructed power relations, and transformed their practice in their projects. Similarly, López-Gopar et al. (2021) detail their implementations of socially just, decolonial English language teaching and assessment practices. Investigation of current issues (n = 5) was also found to be an inquiry focus as a stepping stone for change implementation. The investigations focused on issues such as institutional policies and practices surrounding assessment (Gonzalves, 2017) and teacher awareness of cultural capital in the classroom and its influence on their teaching (Brooke-Garza, 2015). Finally, one study (Ordem, 2023) was categorized as development of pedagogical recommendations; its inquiry focus included the development of a six-stage recommendation for a critical pedagogy-informed syllabus implementation, based on member reflections on the PAR process.
Areas of inquiry of the reviewed studies.
Note. * Study counts are not mutually exclusive.
3.2.3. Criticality
As shown in Table 6, the majority of the studies (n = 13) were deemed to entail elements of criticality in their action component. The critical actions coded here were broadly categorized into three themes: centering students’(teachers’) linguistic/cultural repertoires and knowledge bases (Brooke-Garza, 2015; Lau, 2020; López-Gopar et al., 2021; Modiba & Stewart, 2022; Murillo, 2017; Orsini-Jones, 2023; Song, 2022); drawing on Freirean approaches (Crawford-Garrett et al., 2020; Di Stefano et al., 2021; Ordem, 2023); and focusing on social problems in local communities (Call-Cummings et al., 2019; Coles-Ritchie et al., 2019; Villacañas de Castro, 2017). Six studies were identified as not entailing elements of criticality in their actions (n = 6), with a general focus on improving teaching/assessment effectiveness (De Neve et al., 2022; Doungprom et al., 2016; Gonzalves, 2017; Makruf & Barokah, 2023; Namugenyi, 2019; Puteh-Behak et al., 2015). The other four studies, coded as potentially critical, seemed to show potential signs of critical action; however, this heavily relied on inferential reading due to limited detail on problem statements, objectives, and inquiry foci. The actions coded as potentially critical were about improving teaching/assessment effectiveness (Moloi & Adegoriolu, 2021; Nugent, 2020) and centering students’(teachers’) linguistic/cultural repertoires and knowledge bases (Akello et al., 2016; Kristiawan et al., 2022). It should be noted that three studies coded as non-critical or potentially critical reported the use of critical methodological approaches, adopting a critical qualitative research paradigm (Namugenyi, 2019), critical participatory action research (Nugent, 2020), and critical discourse analysis (Moloi & Adegoriolu, 2021). This may indicate that being methodologically critical is not always aligned with the orientation and characteristics of the action component of PAR, the implications of which I discuss later.
Criticality of actions identified in the reviewed studies.
3.3. The ‘participatory’ component
3.3.1. Group formation
Regarding how the PAR groups were formed (see Table 7), the most frequently coded was purposive sampling (n = 6), in which the authors stated their technique to select teacher participants using terms or phrases such as ‘purposive sampling/sample’ (Doungprom et al., 2016; Kristiawan et al., 2022; Moloi & Adegoriolu, 2021), ‘purposively selected’ (Akello et al., 2016; Namugenyi, 2019), and ‘purposeful sampling’ (De Neve et al., 2022). Four studies were coded and categorized as invitation, in which the process of group formation was explicitly described as involving invitations from academic researchers to teacher participants (Call-Cummings et al., 2019; Modiba & Stewart, 2022; Nugent, 2020; Ordem, 2023). Another four were categorized as post-secondary courses (Coles-Ritchie et al., 2019; Murillo, 2017; Song, 2022; Villacañas de Castro, 2017). In these studies, all teacher participants were pre-service teachers enrolled in the teacher education courses taught by academic researchers, and the course projects were framed as PAR. Two studies (Di Stefano et al., 2021; Gonzalves, 2017) indicated that their projects stemmed from pre-existing collaborative relationships among the PAR members, categorized as built on pre-existing relationships. In three studies, categorized as other, the PAR groups formed through a fellowship program (Crawford-Garrett et al., 2020), an online professional development course (Orsini-Jones, 2023), and a school-university partnership (Makruf & Barokah, 2023). The remaining four studies (Brooke-Garza, 2015; Lau, 2020; López-Gopar et al., 2021; Puteh-Behak et al., 2015) did not explain how the groups were formed, although pre-existing relationships might be assumed. Overall, the majority of the studies (n = 14) involved projects that can be framed as institutional professional development, such as workshops and teacher education programs, where PAR groups were formed through these institutionalized arrangements.
Project group formation in the reviewed studies.
3.3.2. Naming practices used for teacher participants
Table 8 summarizes the naming practices used for teacher participants in the studies. Nearly two-thirds of the studies (n = 15) anonymized teacher participants, using alphabets or numbers (De Neve et al., 2022; Doungprom et al., 2016; Kristiawan et al., 2022; Modiba & Stewart, 2022; Namugenyi, 2019; Ordem, 2023), pseudonyms 2 (Brooke-Garza, 2015; Call-Cummings et al., 2019; Moloi & Adegoriolu, 2021; Nugent, 2020; Puteh-Behak et al., 2015; Song, 2022), or general nouns 3 such as ‘participants’ and ‘instructors’ (Gonzalves, 2017; Murillo, 2017; Orsini-Jones, 2023). No studies provided rationales for their naming practices to anonymize teachers, such as why certain pseudonyms were chosen. For five studies, it was unclear whether the names used for participants were real or pseudonyms (Akello et al., 2016; Coles-Ritchie et al., 2019; Lau, 2020; Makruf & Barokah, 2023; Villacañas de Castro, 2017); it may be assumed that pseudonyms were used, as is a common practice for qualitative research in education. The remaining three used real names, which were determined based on authorial information (Crawford-Garrett et al., 2020; Di Stefano et al., 2021; López-Gopar et al., 2021). Not surprisingly, the coding of naming practices for this review indicated a mutually consequential relationship with the other coding items from the ‘participatory’ component, namely, group formation, data analysis involvement, positionality statements, and participant authorship (e.g. the use of real names both influences and is influenced by how the research is conducted, documented, and authored). Naming practices for participants were thus found to be a telling marker that signifies a form of participation in PAR.
Naming practices used for teacher participants.
3.3.3. Data analysis involvement
As shown in Table 9, the review suggests limited participant involvement in data analysis in PAR. In nearly two-thirds of the studies (n = 15), participants were not involved in data analysis, or their involvement was unclear, as data analysis was explained in the passive voice without specifying agents. In another four studies, some degree of participant engagement in data analysis was confirmed (Brooke-Garza, 2015; Call-Cummings et al., 2019; Modiba & Stewart, 2022; Song, 2022). However, these teachers’ involvement in data analysis was framed as member checking, a data validation method recommended for conventional qualitative research. The remaining four studies indicated or explicitly mentioned that teachers were active participants in data analysis (Crawford-Garrett et al., 2020; Di Stefano et al., 2021; López-Gopar et al., 2021; Puteh-Behak et al., 2015). In Di Stefano et al. (2021), teachers’ involvement in data analysis was described as central to their project; the teachers analyzed data to ‘fulfill the objectives of PAR: planning changes to their practice, and reflecting upon the consequences of these changes’ (p. 761). Crawford-Garrett et al. (2020) elaborated on how collaborative data analysis in PAR contributed to the construction of key themes and categories that constitute their findings. Among these four studies, three used real names for teacher participants, who were also listed as co-authors in the publication (Crawford-Garrett et al., 2020; Di Stefano et al., 2021; López-Gopar et al., 2021).
Involvement of teacher participants in data analysis.
3.3.4. Positionality statements
As shown in Table 10, most studies (n = 21) did not have a section for positionality statements of individual teacher participants. In contrast, two studies, Crawford-Garrett et al. (2020) and Di Stefano et al. (2021), included such positionality statements. In Crawford-Garrett et al. (2020), the positionality statements appear in tandem with an explanation of the research background, which details the Teaching Out Loud project, a critical literacy-informed curricular initiative with five K-12 teachers. Each participant’s positionality is written in a few sentences, with varying mentions of social identities, such as race, ethnicity, gender, sexuality, class, institutional positions, and regional background, as well as the researcher-participant relationship and a brief explanation of their pedagogical focus and orientation. These teachers’ positionality is often highlighted again in the findings section, indicating that it was central to their PAR project, for example: Damon, who identifies as a Gay, Latinx man, felt that when he was a student, school, and classroom spaces consistently denied the validity of his experience as a member of both Gay and Latinx communities. He deliberately drew upon these experiences as he made instructional decisions about how to approach his high school English classroom which serves predominantly Latinx students. (p. 286)
Positionality statement sections in the reviewed studies.
The positionality statements in Di Stefano et al. (2021) also reference social identities such as race, ethnicity, gender, sexuality, institutional positions, and regional backgrounds, along with a description of their pedagogical approaches and their teaching contexts. The statements appear as part of the methodology section, which details participants and settings, and each teacher participant is introduced in a separate paragraph of about five sentences. Like Crawford-Garrett et al. (2020), teacher positionality appears central to their PAR project, which sought to explore how Latiné stereotypes emerge in Spanish language instruction and how to dismantle them. These two studies, which include explicit positionality statements, frame teacher participants as full researchers, using real names and also crediting them as co-authors.
3.3.5. Participant authorship
Table 11 shows whether teacher participants were included as authors in PAR. As indicated above, whether authorship is extended to teacher participants appears to have a strong consequential relationship with the coding results of other items in the ‘participatory’ component. In this review, the coding of authorship mirrors that of naming practices, as I decided whether the names used for participants are real based on authorial information. Consequently, the studies that included the teacher participants as co-authors are Crawford-Garrett et al. (2020), Di Stefano et al. (2021), and López-Gopar et al. (2021). The participants in these studies were positioned as de facto co-researchers, and the coding results of data analysis involvement indicate their active participation in PAR. In addition, the presence of positionality statements signifies the foregrounding of participants’ representations, roles, and engagement in PAR, as mentioned earlier.
Participant authorship in the reviewed studies.
The other 20 studies, where teachers were anonymized or naming practices were unclear, list only academic researchers as authors. The results generally suggest that teachers are unlikely to receive authorship in PAR if they are recruited through purposeful sampling or their registration in post-secondary courses; if they are not involved in data analysis; and if their positionalities are not described as central to the projects. This appears to be closely related to how the ‘research’ component is framed and who is considered a knowledge producer within such framing. I now turn to the results concerning the ‘research’ component of PAR.
3.4. The ‘research’ component
3.4.1. Methodological framing
The majority of the studies (n = 14) explicitly framed their research by drawing on particular PAR theories, citing them in the methodology section or the section that details their research. Seven studies (Call-Cummings et al., 2019; Doungprom et al., 2016; Modiba & Stewart, 2022; Moloi & Adegoriolu, 2021; Namugenyi, 2019; Ordem, 2023; Villacañas de Castro, 2017), categorized as limited, did not cite such theories to explicitly frame their methodological orientations, but they showed some influence of particular theorists when they explained PAR in general, referring to the literature. The remaining two (Gonzalves, 2017; Lau, 2020) did not make references to methodological literature in their PAR description. Table 12 shows these coding results.
Methodological framing in the reviewed studies.
Table 13 shows the methodological theorists cited in more than two studies, suggesting a tendency in citation practices for PAR methodology. While these citation practices indicate that various methodological theories are drawn on in PAR projects, they generally suggest a methodological commitment to viewing PAR as a standalone participatory research tradition, rather than framing projects as a ‘participatory’ version of action research. This is evidenced by the frequent references to Fals Borda, Kemmis, McTaggart, Nixon, and McIntyre, all of whom are PAR theorists. In addition, studies often cited more than one methodological source to frame their PAR projects; for example, López-Gopar et al. (2021) referenced three methodological sources (Fals-Borda, 2001; McIntyre, 2008; Whyte, 1991), and Nugent (2022) referenced four (Elliot, 1991; Kemmis et al., 2014; Lewin, 1946; McIntyre, 2008). Also, citations of literature on other methodologies appear along with those of the PAR methodology, indicating that their PAR projects drew on multiple methodological orientations. Crawford-Garrett et al. (2020) drew from the tradition of practitioner inquiry methodologies such as Cochran-Smith and Lytle (2009) as well as PAR theories. Similarly, Di Stefano et al. (2021) characterized their study design as PAR (McIntyre, 2008) while explaining that it also included elements of collaborative self-study research (O’Dwyer et al., 2019).
Participatory action research (PAR) theories frequently cited in the reviewed studies.
3.4.2. Types of data and data analysis frameworks/methods
Table 14 presents data sources used in the PAR studies. Most of the studies (n = 21) generated and analyzed more than one data source. Data generated by teacher participants via interviews or focus groups were frequently used in the reviewed studies, while data from students were not frequently used. Some may associate this trend with the finding that teachers were not typically treated as de facto co-researchers with authorship, as previously reviewed. However, the use of teacher data does not necessarily imply that teachers were framed as data providers rather than co-researchers. Both Crawford-Garrett et al. (2020) and Di Stefano et al. (2021) included teacher data such as interviews, focus groups, and teaching materials; nevertheless, teachers were involved in analyzing such data and were positioned as co-researchers with authorship. Conversely, the use of student data does not necessarily imply that teachers were actively involved in the research process and were thus framed as co-researchers.
Data sources used in the reviewed studies.
Notes. a Includes project reports and action plans (Crawford-Garrett et al., 2020; De Neve et al., 2022), online discussions (Kristiawan et al., 2022; Nugent, 2020; Song, 2022), project/course artefacts and performances (Kristiawan et al., 2022; Murillo, 2017; Song, 2022; Villacañas de Castro, 2017), professional development session recordings (De Neve et al., 2022; Nugent, 2020). b Includes social media sources and presentation/workshop materials (Di Stefano et al., 2021), student academic achievement (scores) and video/audio recordings (mentioned but not specified) (Doungprom et al., 2016); teachers’ and students’ reflective feedback on focus group discussions (Ordem, 2023). c Includes teacher participants and students.
Regarding the procedural frameworks and methods for analyzing PAR data, general guides and frameworks for qualitative data analysis (e.g. Creswell, 2007, 2012), qualitative coding (Saldaña, 2009, 2013), and thematic analysis (e.g. Braun & Clarke, 2006; Dörnyei, 2007) were frequently used in the PAR studies (for details, see Table 15). Critical approaches to data analysis, such as critical discourse analysis (e.g. Fairclough, 2006; Wodak, 1995) and critical ethnographic analysis (Carspecken, 1996), were adopted in three studies, which indicated the studies’ ideological stance. As shown in Table 15, the reviewed studies reported the use of various data analysis methods employed in other research traditions, while five studies did not mention such data analysis frameworks for PAR.
Data analysis frameworks/methods used in the reviewed studies.
4. Discussion and implications
The findings of this methodological review suggest that PAR has been practiced in diverse ways and for various purposes within additional language education contexts. This reflects the fact that PAR can be conceptualized in various ways, providing a methodological rationale for collaborative research with teachers. The evaluation of these multifaceted conceptualizations and applications of PAR may depend on one’s paradigmatic position and beliefs. Some may consider them as heterogeneity that promotes wider recognition and use of PAR, while others may view some of them as conceptual dilution that risks obscuring PAR’s critical and democratic vision. As I elaborated in my positionality statement, I hold that what is central to PAR is ‘Who is the expert?’ from a critical perspective on the politics of participation. Based on this paradigmatic position, I return to each of the three constructs in PAR to discuss methodological issues and implications. I take this finding, the multiplicity of PAR in action, as an opportunity to reflect on what we do using PAR and how we can better realize its methodological tenets and potential.
4.1. Action in PAR
The findings regarding the ‘action’ component of PAR suggest that the studies generally focused on a particular pedagogical implementation and its evaluation. However, the reviewed studies tended to have varying ideological framings of action in PAR, which may be central to what I described as ‘multifaceted conceptualizations and applications.’ This was highlighted by the coding results of criticality. While just over half of the studies framed action as direct pedagogical interventions to challenge wider social structures and to enact social change and justice, other studies focused on improving teachers’ practices as a form of professional development in subject matter in a rather apolitical and uncritical manner. In this ‘professional development’ PAR, teacher participant groups were often formed through purposeful sampling, signifying the researcher’s centrality in the overall research process. Although action can be conceptualized in multiple ways in PAR depending on the context (Guy et al., 2020), the common thread that makes such action distinct from action in other research traditions is threefold: critical, socially impactful, and participant-driven. This form of action can be distinguished from a technical form of action, such as improving the outcomes of a teacher-researcher’s own practice (e.g. student test scores) (Kemmis et al., 2014), which are often assessed based on positivistic ‘expert’ knowledge (e.g. prescriptive rules, theory, benchmarks). According to Jordan (2009), this technical framing of action can be understood as an ideological orientation in action research as opposed to PAR, and Jordan argues that action research ‘has been principally developed by academic researchers working from universities within the advanced capitalist world of the Global North’ and that ‘its ideological orientation has tended to be liberal, focusing on the improvement of professional practices – this is why it has proven to be so popular among researchers working with teachers and other professional groups’ (p. 17).
The ‘action’ component thus shapes PAR’s ontology, and in turn, has direct implications for enhancing methodological rigor. To fully enact its methodological vision, it would be necessary to pay attention to the ideological framing of action expressed in the project’s goals. After all, all research with teachers entails action; a loose conceptualization of action runs the risk of reducing PAR to a mere methodological label used to frame research with teachers.
4.2. Participation in PAR
The same argument would apply to discussions of the ‘participatory’ component, which is at the heart of PAR (Lenette, 2022). As a principle, PAR invites participants to engage in a bottom-up spiral process in research, not only in the implementation (e.g. teaching) stage but also in the planning and reflection stages (Kemmis et al., 2014; McIntyre, 2008; McTaggart, 1997). That is, participation in PAR is responsive to the question, ‘Who is the expert?’ foregrounding participants’ roles in various stages of research. In this review, however, most studies did not indicate such centering of teacher participants in the research process, as evidenced by their reportedly limited involvement in data analysis and their anonymization. The analysis of group formation also suggests how teacher participants were framed in PAR; ‘sampling’ teachers or forming a group within a post-secondary course may pose a challenge in fostering a bottom-up approach that centers teachers’ agendas (formulating research questions) and reflections based on their interpretation and theorization (data analysis), given the power relations between researchers (teacher educators) and teachers.
Here, some might argue that it is not realistic to expect participants to engage in various stages of research, especially in data analysis, and that it is not uncommon for university researchers to take major responsibility for most of the research process. However, doing so would maintain and reproduce the researcher’s centrality, the very issue that PAR aims to disrupt. At the same time, I recognize that PAR is situated in participants’ real lives under situational conditions and constraints, and that ‘it is unlikely that each party, individually or collectively, can or will participate equally in a PAR process’ (McIntyre, 2008, p. 31). In PAR, people bring to the table different skill sets, knowledge, experiences, resources, and conditions situated and mediated in their real lives. Participation cannot be predetermined or prescribed; it can manifest in various forms across different people and times. McIntyre (2008) thus suggests that a PAR group should have discussions to develop their own understanding of ‘participation’ in research, which allows participation to be ‘viewed as a choice, not an imposition’ (p. 15) of a rigid way of ‘participating’ in research. Yet, this is not to justify participants’ limited involvement in data analysis in PAR; it is integral to the process of collective self-reflection of participants (Kemmis et al., 2014) and should be open to all participants. That is, the framing of data and its analysis in PAR, which I will discuss later, would be critical in discussing the ‘participatory’ component. If data is primarily framed as evidence that only researchers are able to analyze, but not evidence that participants reflect on for further action, the ‘participatory’ component may be diminished in research. Crawford-Garrett et al. (2020), Di Stefano et al. (2021), and López-Gopar et al. (2021) are a few of the studies that demonstrated the centrality of teacher participants in PAR, framing data as evidence for their reflective practice.
The positionality statement was another indicator of how teachers’ participation was framed in PAR. Mentions of participants’ positionality not only in the methodology section but also in other sections would construct the centrality of participants in PAR inquiry and knowledge creation. The foregrounding of participants’ positionality can be taken as a clear marker of PAR epistemologies surrounding the critical and political framing of participation, which are often communicated through phrases like ‘research with people, not on people’ (Merriam & Tisdell, 2016, p. 64) and ‘No research on us without us’ (Fine & Torre, 2019, p. 435). Teacher participants’ social identities inform many aspects of PAR, including research rationales, teaching practices, and discussions, as this review found.
As a methodology that recognizes the centrality and contribution of participants in knowledge creation, participant authorship is an important consideration in PAR. Generally, co-authorship is recognized as a marker of epistemic justice in participatory research (Sarna-Wojcicki et al., 2017). The review identified co-authorship by teacher participants in three studies (Crawford-Garrett et al., 2020; Di Stefano et al., 2021; López-Gopar et al., 2021), suggesting that the more participatory the research, the more likely it is to be co-authored with participants. Beyond naming practices for participants, factors such as participants’ positionality and their involvement in data analysis seemed to shape decisions about co-authorship. As Fine and Torre (2021) explain, the ‘participatory’ component in PAR indeed extends to authorship, where the process of writing, editing, and presenting ideally involves all members of the group. As ethical issues and concerns should be discussed within the group throughout the PAR process (McIntyre, 2008), it is also important to have ongoing conversations about authorial ownership and representation in publications (Fine & Torre, 2021).
Here, I am not suggesting that all participants should be co-authors or use their real names in publications. It is possible that participants prefer not to co-author, opting to remain anonymous for personal and political reasons. Rather, I suggest that issues of co-authorship, representation, and credit be presented as choices and openly discussed between academic researchers and participants in the PAR process. In the reviewed studies that only listed academics as authors, authorship was most likely determined based on direct participation in producing research articles. However, this narrow concept of authorship, as well as the privilege and credence given to writers, should be reconsidered in PAR, whose ontology speaks to the politics of participation, that is, ‘Who is the expert?’ As Sarna-Wojcicki et al. (2017) argue, ‘shifting the meaning of authorship from the physical act of writing to the distribution of authority and accountability in the creation of knowledge is a critical opportunity for rethinking what science is and who can participate in it’ (p. 742). Participatory writing can take the form of ‘You write, and we’ll critique it’ (Fine & Torre, 2021, p. 89); this approach may often be adopted in writing within academic discourse. Realizing the potential envisioned in PAR, therefore, would require explicit and ongoing conversations about authorship and participatory writing, pursuing the ethical and democratic vision of participation in knowledge creation.
4.3. Research in PAR
As the findings regarding the ‘research’ component of PAR showed, most studies cited multiple PAR theorists to frame their methodology, while citing other relevant methodologies. However, it appears that references to such theorists do not always ensure that studies frame their research in ways the cited theories envision. This was particularly highlighted by the analysis of the ‘participatory’ component, as discussed above. The framing of data and its analysis in PAR seems to orient how participatory the projects could be, indicating, in turn, a paradigmatic stance in research. The findings suggest that generated data were often not primarily framed as evidence for participants to reflect on and analyze for further action (Kemmis et al., 2014); rather, they were mostly analyzed by academic researchers using qualitative data analysis frameworks to generate findings, as is common in traditional, non-participatory qualitative research. As indicated by this tendency, the centrality of researchers in data analysis gives rise to two nested layers in PAR: one for the participatory and action components involving teachers and the other for the research component involving researchers, with the former nested within the latter. Alternatively, this two-layered PAR might simply be understood as a case study of PAR rather than PAR itself, foregrounding the role of researchers analyzing a case. The reporting of PAR would thus require considering these paradigmatic layers when deciding how to frame the research (see Figure 6).
Paradigmatic layer(s) of participatory action research (PAR).
4.4. Concluding remarks and reflective questions for PAR practitioners
Focusing on the three components of PAR, this methodological review has examined PAR with teacher participants in the context of additional language education. The goal of this review is not to suggest prescribed guidelines in the form of a to-do checklist for conducting PAR. PAR is indeed a context-sensitive research paradigm, and the forms of its three components are negotiated among all members involved, whose ways of knowing, seeing, and being are situated in the local, across time and space. That said, PAR is not simply a participatory version of action research, with the definition of ‘participatory’ loosely defined. ‘Participatory’ should not be a feel-good adjective to describe collaborative research that sustains epistemic hierarchy among the researcher and the researched; it rather embodies and constructs a self-standing research paradigm that asks ‘Who is the expert?’ – from deciding the research agenda to analyzing data and using the findings. Since PAR is not simply a methodology but rather an epistemology that is reflective about what counts as legitimate knowledge and who benefits from it, its implementation would require not methodological guidelines but instead avoiding being ‘methodological’ in ways that emphasize technical and theoretical aspects of doing research, such as data analysis protocols (Kemmis et al., 2014). That is, academic communities may need to free themselves from preoccupation with what makes research look like research if they are serious about the transformative power and knowledge of participants. It is dialogue among participants that orients and constructs a form of PAR, rather than simply ensuring a generalized protocol. Thus, instead of prescribing and summarizing methodological guidelines of PAR as a conclusion of this review, I would like to offer reflective questions for PAR practitioners, based on the findings (see Table 16). These questions, I hope, will foster and enrich bottom-up dialogue among participants, enabling them to co-construct PAR visions in locally situated ways, which will, in turn, lead to negotiated and mutually agreed social change.
Reflective questions for participatory action research (PAR) practitioners.
Footnotes
Acknowledgements
I would like to thank Dr. Ryuko Kubota, Dr. Meike Wernicke, Dr. Anwar Ahmed, and Dr. Ryan Deschambault for their invaluable feedback on an earlier draft of this article. I would also like to thank the anonymous reviewers for their useful suggestions.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.